| 2 | | Performance tests of our GRAPE-6 emulation library. We have integrated a Plummer sphere for 0.5 timeunit with a blocked timestep scheme. |
| 3 | | We have measured the elapsed time for each model and computed the effective performance of the runs. In all runs, we used 1 GPU board per system. |
| | 2 | Performance tests of our GRAPE-6 emulation library. |
| | 3 | We have integrated a Plummer sphere for 0.5 timeunit with a blocked timestep code. |
| | 4 | We have measured the elapsed time for each model and computed the effective performance of the runs. |
| | 5 | In all runs, we used 1 GPU board per system. |
